SoylentNews is people
SoylentNews
Ovrdrive does not encrypt its contents by default but has a uniquely physical security mechanism and can be rigged to self-destruct - by heating itself to over 100 degrees C:
Through GitHub and Crowd Supply, Ryan Walker of Interrupt Labs (via CNX Software) is releasing a security-focused, open-source USB flash drive called Ovrdrive USB, which boasts a self-destruct mechanism that heats the flash chip to over 100 degrees Celsius.
The Ovrdrive USB is unencrypted by default, so it should still be legal in countries where encryption is otherwise illegal while providing an extra degree of (physical) security not matched by our current best flash drives.
First, the Ovrdrive USB design functions pretty simply. It's mostly a run-of-the-mill USB flash drive with a unique activation mechanism. For it to be detected by your machine, you have to rapidly insert the drive three consecutive times actually to turn it on. Failure to do so will hide the drive's partition and give the impression that it's broken. Initially, it was supposed to self-destruct, but it proved too challenging to mass produce, forcing Walker to change the drive.
[...] In its crowdfunding campaign on Crowd Supply, the flash drive is slated for an August 2024 release and priced at $69 with free US domestic shipping or $12 international shipping for the rest of the world. At the original time of writing, the flash drive has reached 70% of its funding, with two days remaining on the funding deadline.
Related: Report Reveals Decline In Quality Of USB Sticks And MicroSD Cards
(Score: 3, Interesting) by mcgrew on Wednesday February 14 2024, @03:54PM (6 children)
100 degrees Celsius boils water. Heat is the number one enemy of semiconductors. Try taking the battery out of your phone and dropping the phone in a pot of boiling water, taking it out with tongs, and thoroughly drying it over a few days.
I really doubt that after putting the battery back in it will work any more.
Poe's Law [nooze.org] has nothing to do with Edgar Allen Poetry
(Score: 5, Interesting) by owl on Wednesday February 14 2024, @04:21PM (4 children)
Most semiconductors are specified to operate from 70 to 100 degrees C (note, that is operate). Higher grade semiconductors are specified to operate over 100 degrees C.
Lead free solder melt temps begin at 232 degrees C and go up to 250 degrees C. And in the data sheets you'll find specifications for soldering of "X temp for Y time" where the semiconductor (not operating) will tolerate the heat that results from the soldering process (and these temps are substantially higher than 100 degrees C).
If your hypothetical phone quit after this hypothetical boiling water bath, it will not be because 100 degrees C was hotter than the semiconductors could handle, but for some other reason involving water infiltration and damage.
For destruction of a flash memory chip, the heat tolerance of the semiconductor memory chip is the critical component. Boiling water temp hardly seems hot enough to actually damage the chip, so there must be some other method being used, and the marketing dept. got some wires crossed when they wrote "100 degrees C" for the heat level.
(Score: 3, Insightful) by deimtee on Wednesday February 14 2024, @09:36PM (1 child)
TFA is not clear though, is it physically destroying the chip, or is 100C enough to wipe the data on it?
If you cough while drinking cheap red wine it really cleans out your sinuses.
(Score: 2) by owl on Thursday February 15 2024, @10:30PM
A valid point. And while we don't know what brand of flash memory is in this device, here's a data sheet [amazon.com]" for a Micro flash chip. It was simply the first hit from a search for "flash memory chip data sheet".
One has to scroll all the way to pdf page 38 to find the temp ratings, but table 11 says the storage temperature range is -65 °C to +150 °C.
Table 12 (same page) lists the recommended operating conditions and the widest temp range is the "extended" version, specified to operate from -40 °C to +85 °C. Even the "commercial" version has a respectable +70 °C top end recommended.
Given a recommended max of +70 °C or +85 °C (depending on variant) and a storage value of +150 °C, it seems hard to swallow that a measly 100 °C will destroy the chip. Certainly this Micron chip in this datasheet should survive at 100 °C in storage, and would likely operate at 100 °C as well (it is just not guaranteed to actually operate at that temp).
So unless this company has sourced special flash chips that become very leaky at 100 °C (which would make them quite specialized, and expensive) the 100 °C value has the smell of a number the reporter writing the story pulled out of their ass.
And given the fact that so many reporters create "wet streets cause rain [epsilontheory.com]" writeups in their reporting on just about everything, we have to assign a fair probability that this 100 °C is just something conjured out of thin air by the reporter.
(Score: 2) by driverless on Thursday February 15 2024, @02:58AM (1 child)
Yup, 100 degrees won't really do much. A bigger concern though is that this sounds like some sort of pyrotechinc mechanism, which means if he tries to sell it or, worse, ship it, he's going to get a call or visit from some no-nonsense gentlemen who'll inform him about how many laws he'd be breaking by doing so.
OTOH if it's just a nichrome wire or equivalent then I can't see how it's going to be powered, or why anyone would wait around for it to heat the chip up a bit.
Unless there's a lot of detail being missed, this sounds like a gimmick someone thought up after a few beers. The German word schnapsidee [dw.com] springs to mind.
(Score: 2) by driverless on Thursday February 15 2024, @03:01AM
Hit reply too soon, "the mechanism reverses the voltage supplied to the device to around 100 degrees Celsius". Ah, of course, it reverses the polarity of the neutron flow to generate heat.
(Score: 2) by looorg on Wednesday February 14 2024, @04:59PM
Yes that is correct. But it would be some serious flash boiling required, it's not that you go from room temperature to 100C instantly. Over long periods of time operating at 100C is not going to be very good, but here it seems like it's going to go more or less instantly and by that just wreck the circuit or memory chips. In this case it's a specific chip to that needs to get boiled. I'm not sure it's going to work, but I guess he must have figured something out since he is trying to sell people on a crowdfunding effort for it.